We evaluated 5-HT-dependent symptoms as predictors of response to Fluoxetin
e in late life depression (LLD). Thirty-six depressed in- and outpatients a
ged 65 years or more, diagnosed according to DSM-III-R criteria, were evalu
ated using the 21-item Hamilton rating scale for depression (HRSD). After 6
weeks of Fluoxetine treatment (20-40 mg/day), patients were classified as
"responders" (n = 22, > 50 % decrease in the HRSD total score) and "non res
ponders" (n = 14). For the HRSD, both overall scoring and the scores for sp
ecific symptomatologic clusters were analyzed in a statistical model compri
sing the stepwise regression analysis and the exploratory factor analysis w
ith varimax rotation. Among responders, five variables (thought disturbance
, suicide, sleep disturbance, agitation, and obsessive-compulsive symptoms)
accounted for almost 80 % of the total variability. Furthermore, factor an
alysis extrected two factors: the first was dominated by the suicide, sleep
disturbance, and agitation clusters; the second weighted on thought distur
bance and obsessive-compulsive symptoms. As in the younger population, symp
toms related to a 5-HT dysfunction appear to act as predictors of the favor
able response to Fluoxetine in geriatric depression as well.